Inter- and Intraobserver Reliability Between Orthopaedic Surgeons for Reparability of the Anterior Cruciate Ligament Using MRI

Approaching expert-level accuracy for differentiating ACL tear types on MRI with deep learning

Treatment for anterior cruciate ligament (ACL) tears depends on the condition of the ligament. We aimed to identify different tear statuses from preoperative MRI using deep learning-based radiomics with sex and age. We reviewed 862 patients with preoperative MRI scans reflecting ACL status from Hunan Provincial People's Hospital. Based on sagittal proton density-weighted images, a fully automated approach was developed that consisted of a deep learning model for segmenting ACL tissue (ACL-DNet) and a deep learning-based recognizer for ligament status classification (ACL-SNet). The efficacy of the proposed approach was evaluated by using the sensitivity, specificity and area under the receiver operating characteristic curve (AUC) and compared with that of a group of three orthopedists in the holdout test set. The ACL-DNet model yielded a Dice coefficient of 98% ± 6% on the MRI datasets. Our proposed classification model yielded a sensitivity of 97% and a specificity of 97%. In comparison, the sensitivity of alternative models ranged from 84 to 90%, while the specificity was between 86 and 92%. The AUC of the ACL-SNet model was 99%, demonstrating high overall diagnostic accuracy. The diagnostic performance of the clinical experts as reflected in the AUC was 96%, 92% and 88%, respectively. The fully automated model shows potential as a highly reliable and reproducible tool that allows orthopedists to noninvasively identify the ACL status and may aid in optimizing different techniques, such as ACL remnant preservation, for ACL reconstruction.

Medial meniscus tears are most prevalent in type I ACL tears, while type I ACL tears only account for 8% of all ACL tears

PURPOSE

This study aimed to assess the distribution of different anterior cruciate ligament (ACL) tear locations in different magnetic resonance imaging (MRI) planes, and to explore the relationships of ACL tear types with both meniscus injuries and bone bruising.

METHODS

A retrospective study was performed in patients under 60 years old who underwent MRI scans in the sagittal and coronal oblique planes of the knee for ACL tears between 2014 and 2020. Patients with reports of chronic tears, partial tears, or prior surgeries were excluded. Tear locations were classified into five types, and the meniscus tear measurement variables included the presence of ramp, root, bucket-handle, and other types of tears. All injuries were confirmed by arthroscopy. Meanwhile, the presence and location of bone bruising were analysed and scored with the Whole-Organ Magnetic Resonance Imaging Score (WORMS) bone bruising subscale.

RESULTS

A total of 291 patients were included. The prevalence rates of type I and type III injuries were 23/291 (7.9%) and 145/291 (49.8%) in the sagittal plane and 22/291 (7.6%) and 179/291 (61.5%) in the oblique coronal plane, respectively. The prevalence of medial meniscus tears with ACL tears was 126/291 (43.3%), while that of lateral meniscus tears with ACL tears was 77/291 (26.5%). The highest prevalence of medial meniscus injury with ACL tears was 15/22 (68.2%) for type I injuries. Bone bruises were located on the lateral femoral center in 125 patients (46%) and on the lateral tibia posterior in 132 patients (48%); the common areas of bone bruising were slightly correlated with type III ACL tears but not correlated with type I ACL tears.

CONCLUSION

The plane in which an MRI scan is performed affects the classification of ACL tears. The tear type is associated with the prevalence of medial meniscus injuries, and medial meniscus tears are most prevalent in type I ACL tears.

LEVEL OF EVIDENCE

IV.